Compare commits

...

12 Commits

Author SHA1 Message Date
Piyush Singariya
357444c94e Merge branch 'main' into traceop 2026-05-11 20:53:51 +05:30
Piyush Singariya
a8598f3bfa fix: alias all core columns 2026-05-11 20:53:09 +05:30
Piyush Singariya
bca71f9a33 chore: remove comments 2026-05-11 16:04:32 +05:30
Piyush Singariya
c93660357d chore: fmt python 2026-05-11 16:02:18 +05:30
Piyush Singariya
5651e3b7a8 Merge branch 'main' into traceop 2026-05-11 14:28:58 +05:30
Piyush Singariya
cf2cfbc7d4 fix: remove specific of timestamp 2026-05-11 14:27:01 +05:30
Piyush Singariya
a969c38224 chore: fmtlint 2026-05-07 13:53:12 +05:30
Piyush Singariya
b892a0f0a5 chore: file rename 2026-05-07 13:51:22 +05:30
Piyush Singariya
4d47762eba chore: separate e2e test file 2026-05-07 13:50:11 +05:30
Piyush Singariya
77396a0bb3 Merge branch 'main' into traceop 2026-05-07 12:56:59 +05:30
Piyush Singariya
28c05e1bab Merge branch 'main' into traceop 2026-05-04 14:27:19 +05:30
Piyush Singariya
2b9e383994 fix: trace raw export e2e 2026-04-30 15:25:43 +05:30
4 changed files with 665 additions and 46 deletions

View File

@@ -413,28 +413,6 @@ func (b *traceOperatorCTEBuilder) buildFinalQuery(ctx context.Context, selectFro
}
func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFromCTE string) (*qbtypes.Statement, error) {
sb := sqlbuilder.NewSelectBuilder()
// Select core fields
sb.Select(
"timestamp",
"trace_id",
"span_id",
"name",
"duration_nano",
"parent_span_id",
)
selectedFields := map[string]bool{
"timestamp": true,
"trace_id": true,
"span_id": true,
"name": true,
"duration_nano": true,
"parent_span_id": true,
}
// Get keys for selectFields
keySelectors := b.getKeySelectors()
for _, field := range b.operator.SelectFields {
keySelectors = append(keySelectors, &telemetrytypes.FieldKeySelector{
@@ -444,13 +422,38 @@ func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFrom
FieldDataType: field.FieldDataType,
})
}
keys, _, err := b.stmtBuilder.metadataStore.GetKeysMulti(ctx, keySelectors)
if err != nil {
return nil, err
}
// Add selectFields using ColumnExpressionFor since we now have all base table columns
coreFields := []string{"timestamp", "trace_id", "span_id", "name", "duration_nano", "parent_span_id"}
selectedFields := map[string]bool{
"timestamp": true,
"trace_id": true,
"span_id": true,
"name": true,
"duration_nano": true,
"parent_span_id": true,
}
// CH 25.12.5 distributed-analyzer regression (ClickHouse/ClickHouse#103508):
// native columns from a Distributed-table-backed CTE get renamed col → col_0
// in the shard block, making them invisible to the outer SELECT/ORDER BY.
// Fix: rename every core field to a safe alias (_s_<col>) in the inner SELECT
// so the analyzer never sees the original name as an ORDER BY target. The
// outer SELECT re-exposes them under their original names.
innerCoreExprs := make([]string, len(coreFields))
outerCoreExprs := make([]string, len(coreFields))
for i, f := range coreFields {
innerCoreExprs[i] = fmt.Sprintf("%s AS _s_%s", f, f)
outerCoreExprs[i] = fmt.Sprintf("_s_%s AS %s", f, f)
}
innerSB := sqlbuilder.NewSelectBuilder()
innerSB.Select(innerCoreExprs...)
var additionalSelectedFields []string
for _, field := range b.operator.SelectFields {
if selectedFields[field.Name] {
continue
@@ -461,41 +464,56 @@ func (b *traceOperatorCTEBuilder) buildListQuery(ctx context.Context, selectFrom
slog.String("field", field.Name), errors.Attr(err))
continue
}
sb.SelectMore(colExpr)
innerSB.SelectMore(colExpr)
selectedFields[field.Name] = true
additionalSelectedFields = append(additionalSelectedFields, field.Name)
}
sb.From(selectFromCTE)
// Add order by support using ColumnExpressionFor
orderApplied := false
for _, orderBy := range b.operator.Order {
if selectedFields[orderBy.Key.Name] {
continue
}
colExpr, err := b.stmtBuilder.fm.ColumnExpressionFor(ctx, b.start, b.end, &orderBy.Key.TelemetryFieldKey, keys)
if err != nil {
return nil, err
}
sb.OrderBy(fmt.Sprintf("%s %s", colExpr, orderBy.Direction.StringValue()))
orderApplied = true
innerSB.SelectMore(colExpr)
selectedFields[orderBy.Key.Name] = true
}
if !orderApplied {
sb.OrderBy("timestamp DESC")
innerSB.From(selectFromCTE)
innerSQL, innerArgs := innerSB.BuildWithFlavor(sqlbuilder.ClickHouse)
// Outer SELECT: re-exposes core fields under their original names and acts as
// a projection barrier so ORDER BY-only fields do not leak into the result.
outerSB := sqlbuilder.NewSelectBuilder()
outerSB.Select(outerCoreExprs...)
for _, name := range additionalSelectedFields {
outerSB.SelectMore(fmt.Sprintf("`%s`", name))
}
outerSB.From(fmt.Sprintf("(%s) AS t", innerSQL))
if len(b.operator.Order) > 0 {
for _, orderBy := range b.operator.Order {
outerSB.OrderBy(fmt.Sprintf("`%s` %s", orderBy.Key.Name, orderBy.Direction.StringValue()))
}
} else {
outerSB.OrderBy("timestamp DESC")
}
if b.operator.Limit > 0 {
sb.Limit(b.operator.Limit)
outerSB.Limit(b.operator.Limit)
} else {
sb.Limit(100)
outerSB.Limit(100)
}
if b.operator.Offset > 0 {
sb.Offset(b.operator.Offset)
outerSB.Offset(b.operator.Offset)
}
sql, args := sb.BuildWithFlavor(sqlbuilder.ClickHouse)
outerSQL, outerArgs := outerSB.BuildWithFlavor(sqlbuilder.ClickHouse)
return &qbtypes.Statement{
Query: sql,
Args: args,
Query: outerSQL,
Args: append(innerArgs, outerArgs...),
}, nil
}

View File

@@ -67,7 +67,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id, `service.name` FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name` FROM A_DIR_DESC_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -104,7 +104,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_INDIR_DESC_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_INDIR_DESC_B AS (WITH RECURSIVE up AS (SELECT d.trace_id, d.span_id, d.parent_span_id, 0 AS depth FROM B AS d UNION ALL SELECT p.trace_id, p.span_id, p.parent_span_id, up.depth + 1 FROM all_spans AS p JOIN up ON p.trace_id = up.trace_id AND p.span_id = up.parent_span_id WHERE up.depth < 100) SELECT DISTINCT a.* FROM A AS a GLOBAL INNER JOIN (SELECT DISTINCT trace_id, span_id FROM up WHERE depth > 0 ) AS ancestors ON ancestors.trace_id = a.trace_id AND ancestors.span_id = a.span_id) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id FROM A_INDIR_DESC_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "gateway", "%service.name%", "%service.name\":\"gateway%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
@@ -141,7 +141,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_AND_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_AND_B AS (SELECT l.* FROM A AS l INNER JOIN B AS r ON l.trace_id = r.trace_id) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id FROM A_AND_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 15},
},
expectedErr: nil,
@@ -178,7 +178,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_OR_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_OR_B AS (SELECT * FROM A UNION DISTINCT SELECT * FROM B) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id FROM A_OR_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 20},
},
expectedErr: nil,
@@ -215,7 +215,7 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_not_B ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_not_B AS (SELECT l.* FROM A AS l WHERE l.trace_id GLOBAL NOT IN (SELECT DISTINCT trace_id FROM B)) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id FROM A_not_B) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
@@ -380,11 +380,72 @@ func TestTraceOperatorStatementBuilder(t *testing.T) {
},
},
expected: qbtypes.Statement{
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT timestamp, trace_id, span_id, name, duration_nano, parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), __resource_filter_C AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), C AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_C) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_D AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), D AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_D) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), C_DIR_DESC_D AS (SELECT p.* FROM C AS p INNER JOIN D AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id), A_DIR_DESC_B_AND_C_DIR_DESC_D AS (SELECT l.* FROM A_DIR_DESC_B AS l INNER JOIN C_DIR_DESC_D AS r ON l.trace_id = r.trace_id) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id FROM A_DIR_DESC_B_AND_C_DIR_DESC_D) AS t ORDER BY timestamp DESC LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "auth", "%service.name%", "%service.name\":\"auth%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "database", "%service.name%", "%service.name\":\"database%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 5},
},
expectedErr: nil,
},
{
// order-by field (http.request.method) is not present in SelectFields;
// it must be included in the inner SELECT so the outer ORDER BY can
// reference it by alias, but must NOT appear in the outer SELECT list.
name: "order by field not in select fields",
requestType: qbtypes.RequestTypeRaw,
operator: qbtypes.QueryBuilderTraceOperator{
Expression: "A => B",
SelectFields: []telemetrytypes.TelemetryFieldKey{
{
Name: "service.name",
FieldContext: telemetrytypes.FieldContextResource,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
Order: []qbtypes.OrderBy{
{
Key: qbtypes.OrderByKey{
TelemetryFieldKey: telemetrytypes.TelemetryFieldKey{
Name: "http.request.method",
FieldContext: telemetrytypes.FieldContextAttribute,
FieldDataType: telemetrytypes.FieldDataTypeString,
},
},
Direction: qbtypes.OrderDirectionDesc,
},
},
Limit: 10,
},
compositeQuery: &qbtypes.CompositeQuery{
Queries: []qbtypes.QueryEnvelope{
{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Name: "A",
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{
Expression: "service.name = 'frontend'",
},
},
},
{
Type: qbtypes.QueryTypeBuilder,
Spec: qbtypes.QueryBuilderQuery[qbtypes.TraceAggregation]{
Name: "B",
Signal: telemetrytypes.SignalTraces,
Filter: &qbtypes.Filter{
Expression: "service.name = 'backend'",
},
},
},
},
},
expected: qbtypes.Statement{
// http.request.method is in the inner SELECT (so ORDER BY can reach it)
// but is absent from the outer SELECT column list — only the ORDER BY clause references it.
Query: "WITH toDateTime64(1747947419000000000, 9) AS t_from, toDateTime64(1747983448000000000, 9) AS t_to, 1747945619 AS bucket_from, 1747983448 AS bucket_to, all_spans AS (SELECT *, resource_string_service$$name AS `service.name` FROM signoz_traces.distributed_signoz_index_v3 WHERE timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_A AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), A AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_A) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), __resource_filter_B AS (SELECT fingerprint FROM signoz_traces.distributed_traces_v3_resource WHERE (simpleJSONExtractString(labels, 'service.name') = ? AND labels LIKE ? AND labels LIKE ?) AND seen_at_ts_bucket_start >= ? AND seen_at_ts_bucket_start <= ?), B AS (SELECT * FROM signoz_traces.distributed_signoz_index_v3 WHERE resource_fingerprint GLOBAL IN (SELECT fingerprint FROM __resource_filter_B) AND timestamp >= ? AND timestamp < ? AND ts_bucket_start >= ? AND ts_bucket_start <= ?), A_DIR_DESC_B AS (SELECT p.* FROM A AS p INNER JOIN B AS c ON p.trace_id = c.trace_id AND p.span_id = c.parent_span_id) SELECT _s_timestamp AS timestamp, _s_trace_id AS trace_id, _s_span_id AS span_id, _s_name AS name, _s_duration_nano AS duration_nano, _s_parent_span_id AS parent_span_id, `service.name` FROM (SELECT timestamp AS _s_timestamp, trace_id AS _s_trace_id, span_id AS _s_span_id, name AS _s_name, duration_nano AS _s_duration_nano, parent_span_id AS _s_parent_span_id, multiIf(resource.`service.name` IS NOT NULL, resource.`service.name`::String, mapContains(resources_string, 'service.name'), resources_string['service.name'], NULL) AS `service.name`, attributes_string['http.request.method'] AS `http.request.method` FROM A_DIR_DESC_B) AS t ORDER BY `http.request.method` desc LIMIT ? SETTINGS distributed_product_mode='allow', max_memory_usage=10000000000",
Args: []any{"1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "frontend", "%service.name%", "%service.name\":\"frontend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), "backend", "%service.name%", "%service.name\":\"backend%", uint64(1747945619), uint64(1747983448), "1747947419000000000", "1747983448000000000", uint64(1747945619), uint64(1747983448), 10},
},
expectedErr: nil,
},
}
fm := NewFieldMapper()

View File

@@ -72,6 +72,7 @@ class TraceOperatorQuery:
return_spans_from: str
limit: int | None = None
order: list[OrderBy] | None = None
select_fields: list[TelemetryFieldKey] | None = None
def to_dict(self) -> dict:
spec: dict[str, Any] = {
@@ -83,6 +84,8 @@ class TraceOperatorQuery:
spec["limit"] = self.limit
if self.order:
spec["order"] = [o.to_dict() if hasattr(o, "to_dict") else o for o in self.order]
if self.select_fields:
spec["selectFields"] = [f.to_dict() for f in self.select_fields]
return {"type": "builder_trace_operator", "spec": spec}

View File

@@ -0,0 +1,537 @@
"""
Integration tests for TraceOperatorQuery (builder_trace_operator) through the
/api/v5/query_range endpoint.
Covers:
1. Basic trace operator (A => B) — returns matched spans from the correct trace.
2. Order by a field absent from selectFields — must not return a server error.
Guards against the ClickHouse NOT_FOUND_COLUMN_IN_BLOCK regression where
ordering by a column absent from an outer SELECT caused a query failure.
"""
from collections.abc import Callable
from datetime import UTC, datetime, timedelta
from http import HTTPStatus
from fixtures import types
from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
from fixtures.querier import (
OrderBy,
TelemetryFieldKey,
TraceOperatorQuery,
make_query_request,
)
from fixtures.traces import TraceIdGenerator, Traces, TracesKind, TracesStatusCode
def _builder_query(name: str, filter_expr: str, limit: int = 100) -> dict:
return {
"type": "builder_query",
"spec": {
"name": name,
"signal": "traces",
"filter": {"expression": filter_expr},
"limit": limit,
},
}
def test_trace_operator_query_basic(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
Setup:
Insert one parent span and one child span in the same trace.
Tests:
A => B (parent has a direct child) returns the parent span (returnSpansFrom=A)
from the correct trace.
"""
parent_trace_id = TraceIdGenerator.trace_id()
parent_span_id = TraceIdGenerator.span_id()
child_span_id = TraceIdGenerator.span_id()
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=5),
trace_id=parent_trace_id,
span_id=parent_span_id,
parent_span_id="",
name="parent-op",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-a"},
attributes={"operation.type": "parent"},
),
Traces(
timestamp=now - timedelta(seconds=9),
duration=timedelta(seconds=2),
trace_id=parent_trace_id,
span_id=child_span_id,
parent_span_id=parent_span_id,
name="child-op",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-a"},
attributes={"operation.type": "child"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
start_ms = int((now - timedelta(minutes=5)).timestamp() * 1000)
end_ms = int(now.timestamp() * 1000)
response = make_query_request(
signoz,
token,
start_ms=start_ms,
end_ms=end_ms,
request_type="raw",
queries=[
_builder_query("A", "operation.type = 'parent'"),
_builder_query("B", "operation.type = 'child'"),
TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=100,
).to_dict(),
],
)
assert response.status_code == HTTPStatus.OK, response.text
assert response.json()["status"] == "success"
results = response.json()["data"]["data"]["results"]
assert len(results) == 1
rows = results[0].get("rows") or []
assert len(rows) == 1
assert rows[0]["data"]["trace_id"] == parent_trace_id
assert rows[0]["data"]["name"] == "parent-op"
def test_trace_operator_query_order_by_field_not_in_select_fields(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
Setup:
Two traces, each with a grandparent → middle → grandchild chain:
Trace 1: grandparent (svc-a, http.method=POST) → middle → grandchild
Trace 2: grandparent (svc-b, http.method=GET) → middle → grandchild
Tests:
A -> B (indirect descendant) with selectFields=[service.name] and
order=[http.method DESC], where http.method is NOT in selectFields.
1. Query succeeds (no NOT_FOUND_COLUMN_IN_BLOCK error from ClickHouse).
2. Results are actually ordered: POST sorts before GET descending, so
svc-a must come before svc-b.
"""
trace_id_1 = TraceIdGenerator.trace_id()
trace_id_2 = TraceIdGenerator.trace_id()
gp_span_id_1 = TraceIdGenerator.span_id()
mid_span_id_1 = TraceIdGenerator.span_id()
gc_span_id_1 = TraceIdGenerator.span_id()
gp_span_id_2 = TraceIdGenerator.span_id()
mid_span_id_2 = TraceIdGenerator.span_id()
gc_span_id_2 = TraceIdGenerator.span_id()
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_traces(
[
# Trace 1 — grandparent has http.method=POST (sorts first in DESC)
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=5),
trace_id=trace_id_1,
span_id=gp_span_id_1,
parent_span_id="",
name="gp-op",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-a"},
attributes={"operation.type": "grandparent", "http.method": "POST"},
),
Traces(
timestamp=now - timedelta(seconds=9),
duration=timedelta(seconds=3),
trace_id=trace_id_1,
span_id=mid_span_id_1,
parent_span_id=gp_span_id_1,
name="mid-op",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-a"},
attributes={"operation.type": "middle"},
),
Traces(
timestamp=now - timedelta(seconds=8),
duration=timedelta(seconds=1),
trace_id=trace_id_1,
span_id=gc_span_id_1,
parent_span_id=mid_span_id_1,
name="gc-op",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-a"},
attributes={"operation.type": "grandchild"},
),
# Trace 2 — grandparent has http.method=GET (sorts second in DESC)
Traces(
timestamp=now - timedelta(seconds=7),
duration=timedelta(seconds=5),
trace_id=trace_id_2,
span_id=gp_span_id_2,
parent_span_id="",
name="gp-op",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-b"},
attributes={"operation.type": "grandparent", "http.method": "GET"},
),
Traces(
timestamp=now - timedelta(seconds=6),
duration=timedelta(seconds=3),
trace_id=trace_id_2,
span_id=mid_span_id_2,
parent_span_id=gp_span_id_2,
name="mid-op",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-b"},
attributes={"operation.type": "middle"},
),
Traces(
timestamp=now - timedelta(seconds=5),
duration=timedelta(seconds=1),
trace_id=trace_id_2,
span_id=gc_span_id_2,
parent_span_id=mid_span_id_2,
name="gc-op",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-b"},
attributes={"operation.type": "grandchild"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
start_ms = int((now - timedelta(minutes=5)).timestamp() * 1000)
end_ms = int(now.timestamp() * 1000)
response = make_query_request(
signoz,
token,
start_ms=start_ms,
end_ms=end_ms,
request_type="raw",
queries=[
_builder_query("A", "operation.type = 'grandparent'"),
_builder_query("B", "operation.type = 'grandchild'"),
TraceOperatorQuery(
name="C",
expression="A -> B", # indirect descendant
return_spans_from="A",
limit=100,
select_fields=[
TelemetryFieldKey(name="service.name", field_data_type="string", field_context="resource"),
],
order=[
# http.method is intentionally absent from select_fields
OrderBy(
key=TelemetryFieldKey(name="http.method", field_data_type="string", field_context="attribute"),
direction="desc",
),
],
).to_dict(),
],
)
assert response.status_code == HTTPStatus.OK, response.text
assert response.json()["status"] == "success"
results = response.json()["data"]["data"]["results"]
assert len(results) == 1
rows = results[0].get("rows") or []
# Both grandparent spans must be returned
assert len(rows) == 2
# Ordering: POST > GET in DESC — svc-a (POST) must come before svc-b (GET)
assert rows[0]["data"]["service.name"] == "svc-a", f"Expected svc-a (POST) first in http.method DESC order, got {rows[0]['data']['service.name']}"
assert rows[1]["data"]["service.name"] == "svc-b", f"Expected svc-b (GET) second in http.method DESC order, got {rows[1]['data']['service.name']}"
def test_trace_operator_query_order_by_select_field(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
Setup:
Two traces each with a parent → child pair; the parents have distinct durations
(5 s and 1 s).
Tests:
A => B with order=[duration_nano DESC] and no explicit selectFields.
1. Query succeeds (no NOT_FOUND_COLUMN_IN_BLOCK error).
2. Longer-duration parent (5 s) appears before the shorter one (1 s).
"""
trace_id_1 = TraceIdGenerator.trace_id()
trace_id_2 = TraceIdGenerator.trace_id()
parent_span_id_1 = TraceIdGenerator.span_id()
child_span_id_1 = TraceIdGenerator.span_id()
parent_span_id_2 = TraceIdGenerator.span_id()
child_span_id_2 = TraceIdGenerator.span_id()
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=5),
trace_id=trace_id_1,
span_id=parent_span_id_1,
parent_span_id="",
name="parent-long",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-long"},
attributes={"operation.type": "parent"},
),
Traces(
timestamp=now - timedelta(seconds=9),
duration=timedelta(seconds=1),
trace_id=trace_id_1,
span_id=child_span_id_1,
parent_span_id=parent_span_id_1,
name="child-long",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-long"},
attributes={"operation.type": "child"},
),
Traces(
timestamp=now - timedelta(seconds=8),
duration=timedelta(seconds=1),
trace_id=trace_id_2,
span_id=parent_span_id_2,
parent_span_id="",
name="parent-short",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-short"},
attributes={"operation.type": "parent"},
),
Traces(
timestamp=now - timedelta(seconds=7),
duration=timedelta(seconds=1),
trace_id=trace_id_2,
span_id=child_span_id_2,
parent_span_id=parent_span_id_2,
name="child-short",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-short"},
attributes={"operation.type": "child"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
start_ms = int((now - timedelta(minutes=5)).timestamp() * 1000)
end_ms = int(now.timestamp() * 1000)
response = make_query_request(
signoz,
token,
start_ms=start_ms,
end_ms=end_ms,
request_type="raw",
queries=[
_builder_query("A", "operation.type = 'parent'"),
_builder_query("B", "operation.type = 'child'"),
TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=100,
order=[
OrderBy(
key=TelemetryFieldKey(name="duration_nano", field_context="span"),
direction="desc",
),
],
).to_dict(),
],
)
assert response.status_code == HTTPStatus.OK, response.text
assert response.json()["status"] == "success"
results = response.json()["data"]["data"]["results"]
assert len(results) == 1
rows = results[0].get("rows") or []
assert len(rows) == 2
# DESC: 5 s parent first, 1 s parent second
assert rows[0]["data"]["name"] == "parent-long", f"Expected parent-long (5s) first in duration_nano DESC, got {rows[0]['data']['name']}"
assert rows[1]["data"]["name"] == "parent-short", f"Expected parent-short (1s) second in duration_nano DESC, got {rows[1]['data']['name']}"
def test_trace_operator_query_order_by_non_core_field_in_select(
signoz: types.SigNoz,
create_user_admin: None, # pylint: disable=unused-argument
get_token: Callable[[str, str], str],
insert_traces: Callable[[list[Traces]], None],
) -> None:
"""
Setup:
Two traces each with a parent → child pair; parents have distinct http.method
values (POST and GET).
Tests:
A => B with selectFields=[http.method] and order=[http.method DESC].
1. Query succeeds (no NOT_FOUND_COLUMN_IN_BLOCK error).
2. http.method is present in every result row (it is in selectFields).
3. Results are ordered DESC — POST before GET.
"""
trace_id_1 = TraceIdGenerator.trace_id()
trace_id_2 = TraceIdGenerator.trace_id()
parent_span_id_1 = TraceIdGenerator.span_id()
child_span_id_1 = TraceIdGenerator.span_id()
parent_span_id_2 = TraceIdGenerator.span_id()
child_span_id_2 = TraceIdGenerator.span_id()
now = datetime.now(tz=UTC).replace(second=0, microsecond=0)
insert_traces(
[
Traces(
timestamp=now - timedelta(seconds=10),
duration=timedelta(seconds=3),
trace_id=trace_id_1,
span_id=parent_span_id_1,
parent_span_id="",
name="parent-post",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-post"},
attributes={"operation.type": "parent", "http.method": "POST"},
),
Traces(
timestamp=now - timedelta(seconds=9),
duration=timedelta(seconds=1),
trace_id=trace_id_1,
span_id=child_span_id_1,
parent_span_id=parent_span_id_1,
name="child-post",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-post"},
attributes={"operation.type": "child"},
),
Traces(
timestamp=now - timedelta(seconds=8),
duration=timedelta(seconds=3),
trace_id=trace_id_2,
span_id=parent_span_id_2,
parent_span_id="",
name="parent-get",
kind=TracesKind.SPAN_KIND_SERVER,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-get"},
attributes={"operation.type": "parent", "http.method": "GET"},
),
Traces(
timestamp=now - timedelta(seconds=7),
duration=timedelta(seconds=1),
trace_id=trace_id_2,
span_id=child_span_id_2,
parent_span_id=parent_span_id_2,
name="child-get",
kind=TracesKind.SPAN_KIND_INTERNAL,
status_code=TracesStatusCode.STATUS_CODE_OK,
status_message="",
resources={"service.name": "svc-get"},
attributes={"operation.type": "child"},
),
]
)
token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
start_ms = int((now - timedelta(minutes=5)).timestamp() * 1000)
end_ms = int(now.timestamp() * 1000)
http_method_field = TelemetryFieldKey(
name="http.method",
field_data_type="string",
field_context="attribute",
)
response = make_query_request(
signoz,
token,
start_ms=start_ms,
end_ms=end_ms,
request_type="raw",
queries=[
_builder_query("A", "operation.type = 'parent'"),
_builder_query("B", "operation.type = 'child'"),
TraceOperatorQuery(
name="C",
expression="A => B",
return_spans_from="A",
limit=100,
select_fields=[http_method_field],
order=[OrderBy(key=http_method_field, direction="desc")],
).to_dict(),
],
)
assert response.status_code == HTTPStatus.OK, response.text
assert response.json()["status"] == "success"
results = response.json()["data"]["data"]["results"]
assert len(results) == 1
rows = results[0].get("rows") or []
assert len(rows) == 2
# http.method must be present in every row (it is in selectFields)
for row in rows:
assert "http.method" in row["data"], f"http.method missing from row: {row['data']}"
# DESC: POST before GET
assert rows[0]["data"]["http.method"] == "POST", f"Expected POST first in http.method DESC, got {rows[0]['data']['http.method']}"
assert rows[1]["data"]["http.method"] == "GET", f"Expected GET second in http.method DESC, got {rows[1]['data']['http.method']}"